Hi Andrew,
> How heavily loaded are the callback threads cb{Low,Medium,High}?
The two cbLow threads use 5-10% CPU each. Load on the cb{Medium,High}
threads is 0.0%.
> Do you have chains of records that are linked together?
Only a small fraction of the input records are linking to calc records
or similar. No long chains of records or output records that could take
long to process.
> What other resources do your devices share that might limit how fast
> the IOC can communicate with them?
Hmm, I can't think of anything other than RAM (of which we have more
than enough - 24 GB) and the network interface (according to the
hypervisor's monitoring we are using only a small percentage of it's
bandwidth).
> How long does it take to "talk" to each device?
I'll measure the latency tomorrow.
> Multiple callback threads was certainly added, and I suspect it
> might help your situation because the second-half asynchronous
> processing is probably happening there.
I think you're right about this. I actually had to bump up the resources
a little to prevent cbLow queue overflows:
scanOnceSetQueueSize(10000)
callbackSetQueueSize(10000)
callbackParallelThreads(2, "Low")
Queue status after the IOC is running for a while:
epics> callbackQueueStatus
PRIORITY HIGH-WATER MARK ITEMS IN Q Q SIZE % USED Q OVERFLOWS
cbLow 3186 0 10000 0.0 0
cbMedium 0 0 10000 0.0 0
cbHigh 0 0 10000 0.0 0
epics> scanOnceQueueStatus
PRIORITY HIGH-WATER MARK ITEMS IN Q Q SIZE % USED Q OVERFLOWS
scanOnce 630 0 10000 0.0 0
Thanks,
Martin
--
Martin Konrad
Facility for Rare Isotope Beams
Michigan State University
640 South Shaw Lane
East Lansing, MI 48824-1321, USA
Tel. 517-908-7253
Email: [email protected]
- References:
- Excessive scan times for periodic scans Konrad, Martin via Core-talk
- Navigate by Date:
- Prev:
Re: EPICS 7.0.2.1 release preparation Johnson, Andrew N. via Core-talk
- Next:
Re: How to ensure epicsThread function is running in short-lived environment? Benjamin Franksen via Core-talk
- Index:
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
<2019>
2020
2021
2022
2023
2024
- Navigate by Thread:
- Prev:
Re: Excessive scan times for periodic scans Konrad, Martin via Core-talk
- Next:
Re: Excessive scan times for periodic scans Konrad, Martin via Core-talk
- Index:
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
<2019>
2020
2021
2022
2023
2024
|